Fall colors are here!

Winsor Trail, Pecos Wilderness, Santa Fe Basin, Santa Fe, NM.

The encyclopedia of DNA - Part III

The ENCODE project effectively marked the transition from genomics to functional genomics. The goal of the Human Genome Project was to type the entire human genome. Once that was achieved people realized they had just scraped the tip of the iceberg. Today, the goal of functional genomics is go one step beyond DNA sequences, and understand the dynamics of gene expression, transcription, translation and all the complex pathways that lead from DNA to the making of proteins.

In order to do this, the main goal of functional genomics is to annotate regulatory elements of the genome, in other words, elements that regulate gene expression and transcription. For example, proteins called transcription factors bind to regulatory sequences and favor transcription of a gene into mRNA. Last time we learned about regulatory sequences such as promoters and enhancers, and how the ENCODE project has found a vast amount of these sequences, in particular outside and far away from the genes they regulate.

Previously, we also learned about chromatin, the "yarn" of DNA inside the nucleus, and how its configurations affect gene expression. We also learned about transcription factories inside the chromatin, where genes are recruited and transcribed.

In order to transcribe a gene, the two helices of DNA where the gene sits need to be separated. This will allow the RNA polymerase to access the strand where the gene sits and transcribe it. In other words, in order to be expressed, a gene needs to be accessible. To explore how "accessible" a gene is in a specific chromatin configuration, people have employed the technique of mapping regions called hypersensitive sites. These sites are highly accessible to certain enzymes called nucleases, and promoters and most regulatory elements are found in chromatin sites that are hypersensitive to one endonuclease in particular, called DNase I. Therefore, mapping DNase I hypersensitive sites (DHSs) is an efficient way of identifying regulatory DNA regions.

In [1], Thurman et al. identified nearly 2.9 million genome-wide DHSs across 125 cell types.

"Annotating these elements using ENCODE data reveals novel relationships between chromatin accessibility, transcription, DNA methylation and regulatory factor occupancy patterns. [. . .] Patterning of chromatin accessibility at many regulatory regions is organized with dozens to hundreds of co-activated elements, and the transcellular DNase I sensitivity pattern at a given region can predict cell-type-specific functional behaviours."

Chromatin accessibility is what allows transcription factors to bind to the DNA region to be transcribed. Hence, which sites are accessible and which are not plays an important role in gene expression. When transcription factors bind to their target sites, they initiate chromatin remodeling and the recruitment of other chromatin elements. These local perturbations make certain stretches of DNA accessible to nucleases, DNase I in particular.

[1] Robert E. Thurman, Eric Rynes, Richard Humbert, Jeff Vierstra, Matthew T. Maurano, Eric Haugen, Nathan C. Sheffield, & Andrew B. Stergachis, et al. (2012). The accessible chromatin landscape of the human genome Nature DOI: 10.1038/nature11232

The encyclopedia of DNA - Part II

Last week I started discussing the exciting news about the six ENCODE papers published in the Nature September 6 issue. If you haven't already, I highly recommend reading the review ENCODE explained [1], which has a nice summary of the papers and an excellent perspective on what these results mean.

One paragraph in particular is worth quoting:

"The authors report that the space between genes is filled with enhancers (regulatory DNA elements), promoters (the sites at which DNA’s transcription into RNA is initiated) and numerous previously overlooked regions that encode RNA transcripts that are not translated into proteins but might have regulatory roles. Of note, these results show that many DNA variants previously correlated with certain diseases lie within or very near non-coding functional DNA elements, providing new leads for linking genetic variation and disease [1]."

To review what promoters and enhancers are you can take a look at this older post.

I can't stress enough how relevant these findings are. Previously, genes were thought to be the minimal "coding" unit, so much so that the rest of the genome had been dubbed "junk DNA" (and by now you should know how much I hate that unfortunate expression!). In [2], Djebali et al. report that

"about 75% of the genome is transcribed at some point in some cells, and that genes are highly interlaced with overlapping transcripts that are synthesized from both DNA strands [1]."

"The consequent reduction in the length of ‘intergenic regions’ leads to a significant overlapping of neighbouring gene regions and prompts a redefinition of a gene [2]."

Djebali et al. looked at RNA isolates in the whole cell, nucleus and cytosol of 15 different cell lines. They found novel exons, novel splice junctions and sites, and novel transcripts. Many of these elements are in intergenic regions, and many are antisense. They also investigated which of these newly found elements show evidence of protein expression.

When they looked at expression patterns specific to cell lines, they found that gene expression levels were similar across cell lines. The majority of protein-coding genes were expressed across all cell lines, and only a minority (~7%) was specific to certain cell lines. On the other hand, the researchers found many long non-coding RNAs that were largely cell-line specific, while only 10% was expressed across all cell lines. I found this bit to be quite intriguing, as it seems to point that RNAs have a large role in controlling gene expression across cell lines.

Overall, their findings yield an increase overlap in what they call "genic regions". What were previously thought to be "deserts" between genes, aren't so deserted after all, rather, populated by lots and lots of regulatory elements. In their final discussion, Djebali et al. conclude

"The likely continued reduction in the lengths of intergenic regions will steadily lead to the overlap of most genes previously assumed to be distinct genetic loci. This supports and is consistent with earlier observations of a highly interleaved transcribed genome, but more importantly, prompts the reconsideration of the definition of a gene."

[1] Joseph R. Ecker, Wendy A. Bickmore, Inês Barroso, Jonathan K. Pritchard, Yoav Gilad, & & Eran Segal (2012). Genomics: ENCODE explained Nature DOI: 10.1038/489052a

[2] Sarah Djebali, Carrie A. Davis, Angelika Merkel, Alex Dobin,, Timo Lassmann, Ali Mortazavi, Andrea Tanzer, Julien Lagarde, Wei Lin, Felix Schlesinger, & et al. (2012). Landscape of transcription in human cells Nature DOI: 10.1038/nature11233

The encyclopedia of DNA - Part I

The raw numbers of the human genome: three billion base pairs, of which roughly 1% fall into the 20,000 genes in our genome. So, what's all the extra stuff for?

Typing the whole human genome, in 2001, was only the beginning. The next step in disentangling the puzzle was to assign biochemical functions to those three billion base pairs.

"The human genome encodes the blueprint of life, but the function of the vast majority of its nearly three billion bases is unknown. The Encyclopedia of DNA Elements (ENCODE) project has systematically mapped regions of transcription, transcription factor association, chromatin structure and histone modification. These data enabled us to assign biochemical functions for 80% of the genome, in particular outside of the well-studied protein-coding regions" [1].

Let's start with a bit of a refresher.

Regulatory regions: these are regions in the genome that regulate gene transcription. Thanks to these regulatory sequences, skin cells only express "skin" genes, brain cells express "brain" genes, and so on. Promoters, for example, are regulatory sequences found immediately before the start of the gene, on the same strand, and they initiate the transcription of the gene. There are other regions, called enhancer, which also promote transcription. However, contrary to promoters, enhancers need not be near the gene. They don't even need to be on the same chromosome, and some enhancers have been found in introns, regions of a gene that are removed prior to making mRNA.

Transcription factors: I talked a little bit about them last week. These are proteins that can either promote or block the recruitment of RNA polymerase, and therefore either activate or silence a gene.

And, finally you can review the concepts of chromatin structure and histone modification in a few previous posts.

All these concepts are useful to understand that there's a lot, and I mean A LOT going on, between genes and phenotype. Genes are only the starting point. You can't just look at genes alone in order to try and infer a phenotype.

Started in 2003, the aim of ENCODE was to annotate all functional regions of the genome, where by "functional" they don't just mean encoding proteins, but also presenting some biochemical signature such as protein binding or a specific chromatin structure. The latest findings published in Nature: over 700,000 promoter regions and nearly 400,000 enhancer regions that regulate gene expression.

You can see the complications and layers to this: while we have one unique genome, which is identical in all nucleated cells, once you start looking for function, you have to look at the whole genome and chromatin structure and RNA transcripts of all cell lines, as each cell line will have its own activated and silenced genes, its own chromatin signatures, and so on ... whew, that's A LOT!

So far the ENCODE Project Consortium has integrated the data from 1,640 experiments involving 147 different cell types. They saw that

"The vast majority (80.4%) of the human genome participates in at least one biochemical RNA- and/or chromatin-associated event in at least one cell type."

Many more cell lines are yet to be explored, and yet these initial results already shed light into puzzling questions, like, for example: why do nearly 90% of SNPs found in whole genome disease association studies fall outside genes?

"Single nucleotide polymorphisms (SNPs) associated with disease by GWAS are enriched within non-coding functional elements, with a majority residing in or near ENCODE-defined regions that are out- side of protein-coding genes. In many cases, the disease phenotypes can be associated with a specific cell type or transcription factor."

I can't tell you how excited I am about these results, as I started blogging a little over one year ago raising exactly the point that junk DNA should NOT be called junk DNA.

I'm coming down with the flu (how do you explain to your kids NOT to cough in your face when they have a bug? Sigh), so this will be all for this time. But I've got all the Nature papers printed out and will be talking more about them in the next few weeks. A lot of new (and exciting) stuff to learn!

[1] The ENCODE Project Consortium (2012). An integrated encyclopedia of DNA elements in the human genome Nature DOI: 10.1038/nature11247

ENCODE

The current issue of Nature is dedicated to ENCODE, the Encyclopedia of DNA elements. There are six open access articles that discuss the latest findings concerning the vast genomic area that lies between genes. Many of these "junk DNA" regions have indeed a function!

I've been swamped at work, but I'll do my best to read the papers in the next couple of weeks and discuss them here. In the meantime, you can find the papers at www.nature.com, they are open access. Enjoy!

Transcription factories for gene expression: the hard working units of the nucleus

You've probably heard it many times already: if you could stretch out the DNA contained in any one nucleated cell in your body, it would be 2 meters (~6 feet) long. Now imagine packing this 2-meter long molecule into a sphere whose diameter is of the order of a few micrometers, roughly one millionth smaller than a meter. Yes, it's going to be packed in there, yet those genes have to be accessible to the "workers" that come in and perform daily tasks such as gene transcription, replication, and DNA repair. Clearly, which genes are accessible and which aren't is going to play a major role in the cell's life and development.

The chromatin, the ensemble of DNA and proteins inside the nucleus, is dynamically regulated. For gene expression, active genes relocate from chromosome regions and cluster into subnuclear compartments called "transcription factories for gene expression."

As you know, transcription is one of the fundamental steps in the making of proteins: the enzyme RNA polymerase II creates a complementary strand of RNA (a precursor of mRNA) from the active gene. The mRNA is then synthesized and translated into the protein's amino acid sequence. The concept of transcription factories comes from the observation that specific regions in the nucleus are highly enriched in RNA polymerase II, and those are the regions from which new RNA transcripts emerge. A second observation is that distant loci, often on different chromosomes, can interact during regulation through long-range regulatory contacts.

"Increasing numbers of examples suggest that regulatory DNA elements also seem capable of undergoing functional contacts with genes located on other chromosomes. [...] By contrast, temporarily inactive alleles are positioned away from transcription factories, suggesting that genes migrate to these subnuclear sites in order to be transcribed. Crucially, the number of transcription factories per cell is severely limited compared to the number of expressed genes, compelling genes to share the same transcription factory [1]."

The above figure is a schematic of a transcription factory: active genes from different chromosomes are recruited from the chromatin. As transcription proceeds and new RNAs are formed, the templates are reeled through the factory bringing downstream nearby genes. Transcripts generated in a transcription factory that are in close proximity have a greater chance to undergo trans-splicing, in other words, the two transcripts are joined into one even though they originated from different RNA polymerases. The resulting joint RNA is called chimeric RNA. A few studies have observed proteins generated from chimeric RNAs.

In addition to trans-splicing, close proximity in a transcription factory increases the chances of translocation, i.e. one genomic region being moved to a different locus.

"It is puzzling that a genome conformation that increases the risk of potentially grave translocations can evolutionarily persist. We speculate that three- dimensional gene clustering of transcribed loci must elicit evolutionary advantages that outweigh the dangers of translocations."

As Schoenfelder et al. conclude,

"A major challenge will be to decipher the relation between these genome conformation changes and the numerous epigenetic alterations of the genome, allowing their integration into a comprehensive picture of the spatial and functional organization of the nucleus."

[1] Schoenfelder, Stefan, et al. (2010). The transcriptional interactome: gene expression in 3D. Current Opinion in Genetics DOI: 10.1016/j.gde.2010.02.002

How chromatin changes are preserved after cell division

DNA is found in the nucleus of every cell, woven around proteins called histones. This complex of DNA and proteins found inside the nucleus is called chromatin. In the past, I dedicated a couple of posts to chromatin rearrangements, how they are used by the cell to silence certain genes, and how epigenetic reprogramming has to happen in order for cells to differentiate during development. I'm still learning how these epigenetic mechanisms work, and today I'd like to share with you a couple new readings I've done on the topic.

Chromatin complexes that repress transcription during development are formed by a group of proteins called the Polycomb group (PcG). The proteins in this group form two classes, PRC1 and PRC2. From Wikipedia:

"PRC2 is required for initial targeting of genomic region (PRC Response Elements or PRE) to be silenced, while PRC1 is required for stabilizing this silencing and underlies cellular memory of silenced region after cellular differentiation."

In other words, PCR2 recognizes the current chromatin state and targets the regions to be silenced in order to maintain the same state after cell division. This guarantees that an undifferentiated cell like an embryonic stem cell for example, stays undifferentiated for as long as it's needed.

How does PCR2 distinguish active chromatin (activated genes) from the inactivated one (silenced genes)?

Histones are not static. Imagine these molecules undergoing rearrangements every time they need to change the way they interact with DNA. These changes are called histone modifications and are classified based on the type of histone, amino acid, and position at which they undergo the change. Different histone modifications mark different states of the gene. For example, active genes are usually marked by H3K4me3 and H3K36me2/3, whereas inactive genes are marked by H3K27me3.

In [1], Yuan et al. suggest that active genes are not silenced by PRC2 because, besides having the "active" marks, the chromatin region that contains them is also less compact, with a lower density of nucleosomes and histones H1.

"Once active transcription has ceased upon transcription factor dissociation, either the chromatin-remodeling events or the incorporation of additional histones (including linker histones) would lead to higher nucleosome density, higher H1 content, and more compact chromatin structure, which in turn would convert these nucleosomes from their inert status to ideal substrates of PRC2. Thus, H3K27me3 could be established and lead to further repression of the target genes."

To test their hypothesis, Yuan et al. used a mouse model and the gene CYP26a1 as target, and observed that changes in the local density ("compaction") of the chromatin preceded the establishment of silencing marks.

[1] Wen Yuan, Tong Wu, Hang Fu, Chao Dai, Hui Wu, Nan Liu, Xiang Li, Mo Xu, Zhuqiang Zhang, Tianhui Niu, Zhifu Han, Jijie Chai, Xianghong Jasmine Zhou, Shaorong Gao, & Bing Zhu2 (2012). Dense Chromatin Activates Polycomb Repressive Complex 2 to Regulate H3 Lysine 27 Methylation Science DOI: 10.1126/science.1225237

The mystery of the Delta antigen

Now I feel like I should write a spy-fiction story on the delta antigen... Maybe I will... But for now, here's the real story.

When in the mid '70s a group of patients in Turin, Italy, presented a particularly virulent form of Hepatitis B (HBV), medical researchers thought they had found a new subtype of the virus. Liver biopsies from infected patients revealed a new antigen, which was thought to be a new protein encoded by HBV. There was a mystery to solve, though: why was the new antigen (called Delta) found only in certain HBV infected patients but not others?

A collaboration between the University of Turin and the NIH revealed, thanks to experiments conducted on chimpanzee, that the Delta antigen was indeed a new virus, the Hepatitis D virus (HDV). Why was it only found in HBV-infected patients? Because HDV is a satellite virus, in other words, it can only infect a liver cell if the HBV virus is also present in the cell.

Similarly to viroids, the HDV genome is a circular single-stranded RNA and, even though not as small as a viroid genome, with its ~1700 bases, it's still much smaller than the average viral genome. It utilizes the cellular RNA polymerase in order to replicate, and HBV envelope proteins in order to propagate. HDV and HBV coinfection is rare in the western world, but quite common in sub-Saharan Africa, the Middle East, and the northern part of South America, where it is mostly transmitted among drug users (like the other hepatitis viruses, it is a blood-borne disease).

So, how did HDV develop in relation to HBV? Given the close resemblance that HDV has with viroids, Taylor and Pelchat [1] report as a likely hypothesis that it had originated from a plant virus (plant viruses can be disseminated in human digestive tracts) and infected the liver of an HBV-infected animal. It could also have originated directly from HBV through a random replication error. As the researchers conclude,

"Furthermore, discovery of the widespread occurrence of HDV-like ribozymes and their possible relationship to retrotransposition further demonstrates that some of what appeared to be unique properties of HDV and viroids are barely the tip of a major biological iceberg."

[1] John Taylor, & Martin Pelchat (2012). Origin of hepatitis δ virus Future Medicine DOI: 10.2217/fmb.10.15

It's not a virus, it's a viroid

A virus is a stretch of DNA or RNA, usually a few thousand bases long, enclosed in a protein shell. Once inside the cell, the RNA or DNA from the virus starts producing viral proteins, which are then used for replication.

Now imagine a circular strand of RNA that instead of a few thousand bases comprises a few hundred bases. It doesn't code for proteins, it doesn't come in a shell. And yet it's highly pathogenic and able to reproduce. In plants, that is.

A viroid is essentially a circular strand of RNA, typically between ~250 and ~450 bases long, and it doesn't encode for proteins. As a consequence, it depends entirely on cell proteins in order to replicate and propagate. Currently there are 30 known viroids, all belonging to two families, one that replicates in the cell nucleus, and one in plastids (organelles outside the nucleus) instead. While viral infection is prompted by the proteins the virus codes, it still remains a mystery how non-coding viroids can initiate phenotypic changes in their hosts. These changes are broad in degree and extent, with some viroids inducing no changes at all, and others resulting in severe pathogenicity.

"Potential pathways connecting the first viroid-host interaction that through one or more cross-talking signaling cascades ultimately lead to the macroscopic symptoms. Most components of these pathways, including the initial triggering viroid RNA species, are hypothetical [1]."

In [1], Navarro et al. explore various hypothesis as to how viroids initiate infection. One possibility is that viroid-derived RNAs could be targeting host RNA for silencing. They studied one viroid in particular that causes a severe form of albinism in the leaves, stems and fruits of peach seedlings. They deep sequenced healthy and diseased leaves and provided "direct evidence involving RNA silencing in modulation of host gene expression by a viroid."

It turns out, there's a particular human-infecting virus that behaves more like a viroid than a virus, but that story I'll save for next time. :-)

[1] Beatriz Navarroa,, Andreas Giselb,, Maria-Elena Rodioa,, Sonia Delgadoc,, Ricardo Floresc,, & Francesco Di Serio (2012). Viroids: How to infect a host and cause disease without encoding proteins Biochem DOI: 10.1016/j.biochi.2012.02.020

What's that gene for, again?

I'm always skeptical when you hear prepositions such as "gene X has function Y," as often there are very complicated mechanisms nestled between the "gene" and the "function/phenotype." If you've been following me over the past year (yes, I've been blogging for a year already, time flies!), we've learned that between-gene interactions (epistasis), and changes in gene expression (epigenetics) can completely change the picture.

Recent reviews on the use of RNA interference have given me additional reasons to be skeptical.

Gene function in vivo has been studied through a procedure called "gene knockdown," which uses RNA interference (RNAi) to "tune down" the expression of the gene. RNAi has also been used in to mimic human genetic diseases that would otherwise have no somatic equivalent in the animal world, in particular in studies aimed at discovering novel drug targets. By introducing synthetic RNA into the cell, researchers can effectively silence target genes and thus identify their functions within specific cellular processes.

It certainly is a brilliant tool, but there are several issues one needs to keep in mind when using RNAi. The target specificity, for example, is not always perfect, and several off-target effects (down-regulation of genes different from the target ones) have been documented. When this happens, you can no longer be sure of what genes, if not all, caused the observed change in phenotype. Ideally, in order to minimize off-target effects, one should repeat the experiment with different types of RNAi targeting the same gene. Rescuing the loss of function by re-inserting the mRNA (or making it "immune" to the RNAi) would also provide further evidence. However, this is very hard to realize in practice.

It gets more complicated.

Many genes regulate cellular fitness. The change observed change in phenotype, rather than reflect the knockdown gene, could instead be a direct consequence of lower cell proliferation. In addition, we tend to simplify things thinking that the relationship between gene and phenotype is linear, or that the effect from different genes is additive, when in fact such simple mathematical frameworks often don't capture the reality of the biological world. Interactions and non-linearity are difficult to model. Experiments that target multiple genes rank the results in terms of dose-responses, though such results are often contaminated by false positives and knockdown efficiencies.

All this not to say that this is the end of RNAi experiments, rather, that additional thought has to be given when interpreting the results. We still have a long way to go before we can fully encompass the complexity of our genome, and we are taking one baby step at the time.

William G. Kaelin Jr. (2012). Use and Abuse of RNAi to Study Mammalian Gene Function Science, 337 (6093), 421-422 DOI: 10.1126/science.1225787

Olympic fever, olympic medicine

Will you be missing the Olympics now that they are finally over? I will.

If you enjoyed watching the games, you'll also enjoy the NEJM perspective by David Jones [1], which gives a historical overview of olympic medicine and how it has studied, over the years, the limits of the human body.

It's interesting to read how in the 1904 games in St. Louis the marathon winner had taken strychnine sulfate, five eggs, and brandy during the race and still required medical attention afterwards. Fast forward to the 1984 Olympics in Los Angeles: seven cyclist received blood transfusion to enhance their performance, a practice later condemned by the Olympic Committee.

What drives us to push the limits? It's indeed spectacular to see how far the human body can do and what it can achieve, and yet we step back horrified when we hear about performance-enhancing drugs, or when we read about the borderline training young gymnast undergo (somebody went as far as to define it "child abuse"). When does the body stop being a body and when does it start being a machine? Some sports have in the past received harsh criticism for the life-long consequences they carry. Is medicine's role to let us enjoy these beautiful performances while preserving the athlete's health, or is it to keep pushing the envelope and see how far we can go? Where do we draw the line between what is allowed and what is not? If athletes went as far as blood transfusions in 1984, does it mean gene therapy will be the new dare in the next decade? Would you be willing to permanently alter your genes in order to reach your dreams?

[1] David S. Jones (2012). Olympic Medicine New England Journal of Medicine, 367, 289-292

Rainbow

This was last Tuesday, August 7th. Presumably the largest rainbow in a while, it made it to the local paper.

Anderson Overlook

Not much time to read papers lately, I apologize.

So I give you images instead. This is my favorite spot to shoot at sunset this time of the year. The sun hits it just at the right angle, and there's often stormy clouds sweeping through, which make the whole scene more interesting. It's called Anderson Overlook and those formations you see are part of the Bandelier National Monument.

Feel free to tell me which ones are your favorites. The local library has a call for landscape photographs of the area and I'm thinking of submitting one or two from this set.

The beginning of the end. . . Maybe.

"We share a very special moment - it is the moment when an AIDS-free generation is finally in sight." That's what the US president, Barack Obama, said on July 26.

Well, are we?

A colleague a few days ago brought to our attention some stunning figures: according to the CDC, of all HIV infected individuals in the US, only 25% are under treatment and hence have the virus under control. Quite striking if you consider that in Sweden instead 85% of HIV-positive individuals are undergoing treatment. The consequences of such a poor statistic in the US go beyond the lifespan of the single individual: people under antiretroviral therapy have much lower viral loads and therefore a significantly reduced chance of passing the virus to their partners.

I'm not surprised by the CDC numbers, actually. When newly infected, subjects have no symptoms or may feel like they are coming down with the flu. You can live with this virus for ten years without having symptoms. If you don't have health insurance, and you are feeling well, why bother go see a doctor? In the meantime, though, these individuals continue to spread the virus. And the problem doesn't affect the US alone: according to the World Health Organization, less than half of the infected people worldwide are actually receiving treatment.

These were my thoughts as I read the perspective article "The beginning of the end of AIDS?" in NEJM [1]. The authors base their cautious optimism on a few things: mildly positive results on a recent vaccine trial, more effective drugs, and the news of the first patient to ever be cured of HIV. The latter I discussed in this post. The news was indeed exceptional but, unfortunately, gene therapy is not the way to stop this pandemic: 2/3 of people currently living with HIV/AIDS are in sub-saharan Africa, where drugs are still hard to find, let alone extremely costly procedures like gene therapy. And more effective drugs are not going to solve the problem if they remain unaffordable or unavailable to the majority of infected people.

So yes, in the end, it all boils down to funding:

"Global resources have been declining, not growing, in this period of scientific success. This lack of funding is the major point of divergence between optimism and pessimism."

Why invest on HIV?

"Comprehensive economic models predict that making the needed investments in HIV-related efforts will result in cost savings over the long term."

HIV debilitates the immune system. The effects of diseases like tuberculosis, hepatitis and malaria could be reduced if the spread of HIV could be reversed because of the effect the virus has on the immune system. Making antiretroviral treatment available to all infected people is the best strategy: by keeping the viral load under control, drugs effectively lower the rates of mother-to-infant and sexual infections. According to the CDC, about 100-200 infants are born every year in the US with the virus in their body. Adequate treatment during pregnancy and delivery can reduce the rate of mother-to-child transmission to less than 2%.

As we strive to reach out to every infected person on the planet, funding must not stop for research. A vaccine is the most affordable and most effective way to stop the pandemic and we have to keep pushing in that direction. My supervisor gave a talk last week in which she outlined where we are in terms of vaccine research. She opened the talk remembering how the first vaccine was discovered: the English physician Edward Jenner (1749 - 1823) rubbed pus collected from blisters milkmaids received from cowpox on his gardner's eight-year-old son. He then exposed the boy to pox, twice, and noticed that they boy didn't develop the disease. The audience was of course horrified when my supervisor mentioned the sacrifice of the little boy, and yet when she went on describing how long and, most importantly, how much money is needed to develop and test a vaccine, a girl in the audience raised her hand and asked: "Well, maybe you can't take an eight-year-old boy, but wouldn't you be better off testing the vaccine on yourself?"

We can't, of course. The gardner's boy got lucky, but things don't always go well. The field still hasn't forgotten the failure of the Merck vaccine in 2007, the trial that was halted after the experimental vaccine was found to make some subjects more susceptible to infection. The FDA has a set of very strict regulations on vaccines. They need to be stable, in other words, one has to show that they don't change after they've been grown for several generations in cultures. They have to be attenuated, and remain so after several generations. Vaccines are then tested on mice, first, then monkeys, then, years later, on humans in several phases that take time, money, and then more time and more money. And yet what we really cannot afford is to stop pushing the research forward.

"Every country must develop more effective ways to reach key affected populations and to apply the tools that we know work, if we are to make significant advances."

So, Mr. President, I hope you are right. But I also hope you will keep funding our efforts.

[1] Diane Havlir, & Chris Beyrer (2012). The Beginning of the End of AIDS? New England Journal of Medicine : 10.1056/NEJMp1207138

Oedipus's dilemma

I love Greek mythology, and of all myths, Oedipus is probably the one that fascinates me the most. Nothing to do with the fact that it's become a psychiatric hallmark. I love this myth because it always makes me wonder: if somebody came to you and told you they knew with absolute certainty your future (how many years you'll live, what you'll accomplish, etc.), would you want to know? It's a paradox, because that knowledge would affect the future course of action you choose. Think about Laius: he fulfilled his destiny exactly because of the actions he took in order to avoid his destiny. Predestination paradoxes have been used forever in all mythologies, and even these days -- can you think of at least a novel or a movie where it's been used?

I'm rambling, but I actually have a point for this post, I promise.

As you know, nobody's going to come and offer to tell you your exact destiny. But, they might offer to type your entire genome. And from that, they may argue they can tell you the exact risk you have of developing certain diseases. In fact, some of you may already have opted to have their entire genome typed. Such services have become more affordable, accurate, and efficient in just a handful of years. The benefits are numerous: drug therapy could be genetically targeted, and just by looking at your DNA your doctor could already know which drugs will be more effective and which could instead have adverse effects. Assessing one's risk for cancer, diabetes, or other diseases can be a good motivator to a healthier lifestyle and open up preventive treatment choices.

So, where's the catch?

The catch is that, as a new study on Science Translational Medicine shows [1], sequencing the entire genome doesn't tell us the whole story. In fact, in many cases, it doesn't tell us much at all.

Roberts et al. argue that the risk we need to be able to assess should be pretty strong in order to make preventive measures effective. For example, currently the general population risk of developing breast cancer within a woman's lifetime is 12%, obviously too low for women to opt for a preventive mastectomy. However, if a woman learned that her risk was 90%, she might reconsider. Any preventive measure carries consequences, and therefore, the risk reduction it ensures should be pretty strong in order to establish clinical utility.

After setting a meaningful risk threshold, Roberts et al. collected genetic data from numerous homozygous twin registries and cohorts. (Little pet peeve of mine: couldn't find the exact number of pairs they had in the study, it's probably in the supplemental material, but I find sample size important enough to expect it in the main text). They then developed a mathematical model to estimate the maximum capacity of whole-genome sequencing to predict the risk for 24 common diseases, including autoimmune diseases, cancer, cardiovascular diseases, genito-urinary diseases, neurological diseases, and obesity-associated diseases. The idea behind the mathematical model is to assess the risk increment of an individual with a disease-associated genotype compared to someone with no genetic risk at all. Since homozygous twins have nearly identical genomes, you would expect their genetic risks to have a nearly identical outcome.

"The general public does not appear to be aware that, despite their very similar height and appearance, monozygotic twins in general do not always develop or die from the same maladies. This basic observation, that monozygotic twins of a pair are not always afflicted by the same maladies, combined with extensive epidemiologic studies of twins and statistical modeling, allows us to estimate upper and lower bounds of the predictive value of whole-genome sequencing."

Using their model, the researchers showed that most individuals would show a risk predisposition to at least one of the 24 diseases tested. At the same time, they would test negative for most diseases. What does this mean? It means that we cannot predict the risk allele distribution of the actual population, and most often genetic testing will only say that individual X has the same risk of developing disease Y as the general population -- hardly enough to make whole genome testing surpass the clinical utility threshold.

"Thus, our results suggest that genetic testing, at its best, will not be the dominant determinant of patient care and will not be a substitute for preventative medicine strategies incorporating routine checkups and risk management based on the history, physical status, and life-style of the patient."

[1] Nicholas J. Roberts, Joshua T. Vogelstein, Giovanni Parmigiani, Kenneth W. Kinzler, Bert Vogelstein1 and, & Victor E. Velculescu (2012). The Predictive Capacity of Personal Genome Sequencing Sci Transl Med 4, 133ra58 DOI: 10.1126/scitranslmed.3003380

The vulnerable banana crop

"What are you doing?"
"Eating a banana."
"Did you know that banana trees are seedless? They only reproduce asexually and hence are all genetically identical."
(Me, chewing) "Hmm-mmm."
"If a parasite were to kill one, it would kill all of them because there's no genetic variation among the plants."

See, this is what you get from growing up with a biologist father. Over a meal, you can learn infinitely many new things, like the fact that shellfish is an unfortunate name for something that really isn't a fish. The story behind bananas, though, is fascinating. Between 8,000 and 7,000 years ago humans started selecting and hybridizing a number of banana tree species, which eventually lead to the creation of the domesticated banana tree we know today. For the most part, they derived from two species, Musa acuminata and Musa balbisiana. About half of the global banana production comes from this two species. A recent study published in Nature by D'Hont et al. examined the whole genome of the Musa acuminata and reconstructed the history of its domestication through phylogenetic analysis [1].

These plants are mostly triploid, meaning they have three copies of each chromosome. Most sexually reproducing organisms have two copies, and are hence called diploid. A whole genome duplication happens when an organism inherits an additional copy of the entire genome. Triploidism is mostly observed in plants, and often artificially sought to create seedless fruits because triploid organisms are usually sterile. In fact, banana trees are propagated by replanting cuttings. This of course cuts many opportunities for genetic variation. The species ends up being genetically homogeneous, which means that any potential threat to one organism, will be a threat to the whole species. There isn't enough variation to grant a fitness advantage of a subgroup over the other individuals.

As D'Hont et al. conclude in their Nature Letter,

"The reference Musa genome sequence represents a major advance in the quest to unravel the complex genetics of this vital crop, whose breeding is particularly challenging. Having access to the entire Musa gene repertoire is a key to identifying genes responsible for important agronomic characters, such as fruit quality and pest resistance."

Angélique D’Hont,, France Denoeud,, Jean-Marc Aury,, Franc-Christophe Baurens,, Françoise Carreel,, Olivier Garsmeur,, Benjamin Noel,, Stéphanie Bocs,, Gaëtan Droc,, Mathieu Rouard,, Corinne Da Silva,, & et al. (2012). The banana (Musa acuminata) genome and the evolution of monocotyledonous plants Nature DOI: 10.1038/nature11241

An interplay between GO genes and STOP genes optimizes cancer growth

Tumor cells typically have certain genes (called oncogenes) that are either mutated or highly expressed (for example, they can increase in copy number) and that promote tumor growth. Oncogenes often act in combination with silenced tumor suppressor genes -- genes that inhibit tumor development. As the name suggests, if both copies of a tumor suppressor gene are silenced, tumor growth is promoted.

To date, there are nearly 500 oncogenes that have been catalogued and whose mutations have been shown to cause cancer. However, when researchers look at the whole genome of a cancer cell, they find thousands of mutations, the vast majority of which does not affect the cancer genes. Some genes present mutations that clearly promote tumorigenesis, but most mutations found in cancer genomes seem to have a cumulative effect on the proliferation of the cancer.

For example, one common change found in tumor cells is the absence of entire DNA loci. Some deletions happen on both chromosome copies, but most common are the ones that happen on one of the copies only. These are called hemizygous deletions and can span up to thousands of genes. They have been found in breast, gastric, bladder, pancreatic, and ovarian cancers, all with an average of more than ten deletions per tumor. One hypothesis of why they are more common than homozygous deletions is that cells where both copies are deleted are more likely to trigger cell apoptosis and other "self-correcting" mechanisms. Also, in some cases, there may be genes nearby that cannot have both copies deleted.

In order to understand the role of these hemizygous deletions in tumorigenesis, Solimini et al. [1] studied the mutations from whole-genome sequencing of 526 tumors in the Catalogue of Somatic Mutations in Cancer (COSMIC). In their model, they called the tumor suppressor genes STOP genes (genes that inhibit tumor proliferation), and the oncogenes GO genes (genes that promote tumor growth). They found that the majority of deletions were indeed hemizygous and that haploinsufficiency (the loss of one gene copy) of both GO and STOP genes caused by hemizygous deletions is one of the driving forces for cancer growth, and that the effect was cumulative in the number of genes that displayed haploinsufficiency.

A better understanding of these mechanisms may help us target cancer treatment more efficiently.

[1] Nicole L. Solimini,, Qikai Xu,, Craig H. Merme,, Anthony C. Liang,, Michael R. Schlabach,, Ji Luo,, Anna E. Burrows,, Anthony N. Anselmo,, Andrea L. Bredemeyer,, Mamie Z. Li,, Rameen Beroukhim,, Matthew Meyerson,, & Stephen J. Elledge1 (2012). Recurrent Hemizygous Deletions in Cancers May Optimize Proliferative Potential Science DOI: 10.1126/science.1219580

Stress-induced epigenetic changes last up to four generations in mice

One of the most intriguing aspects of epigenetics is its ability to confer transgenerational changes. General belief used to be that inheritance pertained exclusively to DNA, and that what did not affect DNA could not be inherited. Epigenetics encompasses all molecular "processes that regulate genome activity independent of DNA sequence [1]." It has revolutionized the way we view heritability: epigenetic changes do not alter the DNA sequence, only the way genes are expressed. And yet environmental exposures and chronic stress, two factors that can indeed change gene expression patterns, have been shown to induce epigenetic transgenerational inheritance. In other words, you could have inherited some epigenetic switch from your parents, even though the epigenetic switch was caused by some exposure your parents experienced, not you!

In order for this to happen, the epigenetic modification has to be incorporated into the germ line.

In a recent PNAS paper [1], Crews et al. showed the

"epigenetic transgenerational inheritance of a behavioral phenotype induced by an environmental toxicant (a fungicide) and transmitted through the germ line, involving a permanent alteration in the sperm epigenome (i.e., DNA methylation)."

Crews et al. looked at the effects of chronic restraint stress in young male mice. Because social status also influences the way individuals react to stress, with dominant individuals usually being able to cope better than subordinate ones, they housed the experimental mice together with different mixes of social structures. The "stress" was the exposure of a gestational female to a fungicide (vinclozolin), which disrupts endocrine activity. The effects were changes in the brain and behavior and, eventually, the early onset of disease. These were still observed over four generations later.

"We find that this ancestral exposure promotes weight gain and, as such, provides pivotal empirical evidence that exposure to an endocrine disruptor in generations past results in substantial weight gain in the descendants."

In the study, the authors refer the exposure of the mother as "ancestral exposure" to indicate that it wasn't a direct exposure on the individuals under study.

In particular, the researchers observed that the changes in body weight were correlated with lower secretions of corticosterone and higher testosterone circulating levels. The researchers performed other tests in order to measure the sociability of the fungicide exposed animals versus the non-(ancestrally)-exposed ones under stressful circumstances. In both stressful and non-stressful circumstances, the animals ancestrally exposed to the fungicide showed higher levels of anxiety.

Finally, Crews et al. performed gene networks analyses in order to evaluate changes in gene expression between the two mice groups. They looked in several brain regions, including subregions of the hippocampus and the primary and secondary motor cortex. Interestingly, the most altered pathway was the olfactory one.

"An olfactory receptor promoter has been shown to have an epigenetic transgenerational alteration in sperm. [...] Why should genes involved in olfaction be expressed in areas of the brain not involved with olfaction and taste? Olfactory and vomeronasal receptors as a group are among the most rapidly evolving of all genes and have been linked to higher processing centers in the brain as well as to behavior."

[1] David Crewsa, Ross Gillettea, Samuel V. Scarpinoa, Mohan Manikkamb, Marina I. Savenkovab, and Michael K. Skinner (2012). Epigenetic transgenerational inheritance of altered stress responses PNAS DOI: 10.1073/pnas.1118514109